Flu Season Analysis Project

3 Analysis Overview

I have been hired to assist a staffing agency that provides temporary workers to clinics and hospitals on as as needed basis. The agency needed help planning for influenza season as they have limited resources. The U.S. has an influenza season where more people suffer than usual. Hospitals and clinics need this extra staff to treat many of these patients.

This project consisted of an analyzation of factors that contribute to influenza levels nationally. The analysis and business requirements were used to formulate research questions that were later tested.

Data:

The data for this project consists of an influenza deaths data set acquired from the CDC, a population data set from the U.S. census bureau, influenza laboratory test results data from the CDC, and survey flu shot rates in children from the CDC. The data was analyzed in Excel and Tableau.

Skills/Tools

Excel
Tableau
Data Integration/ Transformation
Statistical Hypothesis Testing
Forecasting
Verbal Presentation
Visual Analysis

Project Assumptions

Vulnerable populations suffer most severe impacts from flu and are the most likely to end up in the hospital.
Flu shots decrease the chance of becoming infected with the flu.

Project Constraints

The staffing agency has a limited number of nurses, physician assistants, and doctors on staff.
There’s no money to hire additional medical personnel

What differences can you find in ordering habits of different customer profiles?

Project Steps

1

This was a two part project. In part one, I began to design a data research project, ask relevant questions, and describe the data sets I had access to. The hypothesis was outlined: people aged 65 or older will have more percent positive influenza lab test results than people age lower than 65. I summarized the data sets and began cleaning relevant data in Excel. This cleaned data was compiled into an excel report that can be viewed here. Figure 1 is an example of defining variables listed in the report.

Figure 1

3

This next step was the beginning of part two of my project. Here, I looked to use Tableau to visualize my data. I generated composition/ comparison charts, temporal visuals and began looking at forecasting my data. All of these visuals are viewable on my tableau public via the link on the introduction page. I was able to answer some research questions based on the surplus of visuals created in this step. I used histograms to look at the distribution of ages in influenza deaths, created scatter plots, and bubble charts. Below, in figure 3, you can see various charts created to display this data in Tableau.

Figure 3

2

Now that I had my data outlined, I started to integrate it using excel, based on key columns. Once integrated, I calculated the standard deviation of key variables . I tested the correlation of my hypothesis and interpreted the results. I created an interim report, consolidating my findings to present to the stakeholders. This report can be viewed here. Figure 2 is an image of the t-test conducted on my hypothesis.

2

Figure 2

4

In the final step of this project, I worked to compile a spatial and Textual analysis in Tableau. Both can be viewed on my Tableau Public page. Once completed, I began to compile the results into a final presentation to stakeholders through a Tableau Storyboard which can be viewed here. The final analysis includes my recommendations, interactive visuals, and the relevant charts created throughout the project. Below, my final presentation uploaded to YouTube is visible for viewing.

Final Results

My findings are packaged within the final storyboard link or video. Listed below are what comprises the presentation:

introduction
When additional staff should be sent
Hypothesis test results
Factors to consider

I found that influenza season is the worst in the winter months (December-March). There is a strong relationship between influenza deaths and elderly populations. California suffered the most from influenza and has the highest vulnerable population (aged 65+).

Recommendations

I recommended that the staffing agency send more staff to California as well as be prepared for worse flu number from December through March. They should also send additional staff to areas with high vulnerable populations such as California, Texas, and Florida.

Challenges

I faced several challenges throughout this project. Prior to the project, I had never used Tableau before. It took a while to feel comfortable with the software. Since I do not have a premium subscription to Tableau, I was required to use the Tableau Public software. Due to a hardware issue, my Tableau public would not load my data set. This took some time, but in the end, I was able to use a free trial after doing some research.

In the future, I plan to continue to practice my Tableau skills. As with anything, it takes practice to be good. I very much enjoyed creating visuals in Tableau, and would love to improve my abilities in making them more advanced and interactive.

Previous Project

Next Project